A Bloom's deficient mouse model (Blmm3/m3) has been shown to induce colorectal tumorigenesis when crossed with Apc+/Min mice. Here, we investigated whether the Blmm3/m3 genotype could induce tumorigenesis in extracolonic tissues in tuberous sclerosis 1–deficient (Tsc1+/−) mice that are predisposed to renal cystadenomas and carcinomas. Genotyping of offspring from Tsc1+/− Blm+/m3 intercrosses showed that a ∼24% excess of Tsc1+/− over Tsc1+/+ mice died before weaning (P = 0.016), although Blm deficiency had no cumulative effect on Tsc1+/− survival. Tsc1+/− Blmm3/m3 mice had significantly more macroscopic and microscopic renal lesions at 3 to 6 months compared with Tsc1+/− Blm+/m3 mice (P =0.0003 and 0.0203, respectively), and their tumors showed significantly increased levels of somatic loss of heterozygosity (LOH) of the wild-type Tsc1 (Tsc1wt) allele compared with those from Tsc1+/− Blm+/+ mice (P < 0.0001). Tsc1+/− Blm+/m3 mice did not show significantly more renal lesions compared with Tsc1+/− Blm+/+ animals; however, their lesions still showed significantly increased levels of somatic LOH of the Tsc1wt allele (P = 0.03). Ninety-five percent (19 of 20) of lesions from Tsc1+/− Blm+/m3 mice retained the wild-type Blm (Blmwt) allele, indicating that the increased somatic LOH at Tsc1 was mediated by Blm haploinsufficiency. Renal lesions from a Blm-deficient background stained positively with anti-phospho-S6 ribosomal protein (Ser240/244), suggesting that these lesions develop through the normal pathway of Tsc-associated tumorigenesis. This work shows the use of the Blmm3/m3 mice for inducing renal tumorigenesis, and the high levels (∼87%) of LOH in the resultant tumors will help facilitate mapping of loci involved in tumor progression.

Bloom syndrome is a unique hereditary cancer predisposition disorder in which patients are predisposed to the development of many cancer types (1). The Bloom syndrome gene, BLM, encodes an ATP-dependent, 3′ to 5′ helicase with homology to RecQ DNA helicase (2). BLM resolves Holliday junctions, suppresses recombination in vitro, and is required for the fidelity of DNA double-strand break repair (35). Loss of BLM activity causes genomic instability characterized by an elevated frequency of the sister chromatid exchange (69). Recently, several investigators have exploited the high rate of mitotic recombination in Blm-deficient embryonic stem cells to allow powerful recessive genetic screens to be carried out in vitro (10, 11). Furthermore, Luo et al. (12) have generated a mouse model of the human syndrome (Blmm3/m3) that is prone to a wide variety of cancers and can been used to induce colorectal tumorigenesis when crossed onto an Apc+/Min background. Here, we investigated whether Blmm3/m3 mice could induce tumorigenesis in extracolonic tissue by crossing with tuberous sclerosis 1–deficient (Tsc1+/−) mice that are predisposed to renal cystadenomas and carcinomas (13). Renal tumors develop in Tsc1+/− mice according to Knudson's “two-hit” hypothesis with somatic inactivation of the Tsc1wt allele required for tumor development (13). Normally, a diverse range of mechanisms (including nonsense, frameshift and deletion mutations, and epigenetic silencing) are implicated in the somatic inactivation of the wild-type allele of tumor suppressor genes, and this diversity of somatic alterations makes genetic mapping of loci involved in tumor progression problematic. We hypothesized that by crossing Tsc1+/− mice onto a Blm-deficient background, we would increase the frequency of somatic loss of heterozygosity (LOH), thereby providing a convenient means of directly mapping genes involved in Tsc-associated tumorigenesis.

Animal care, genotyping, necropsy, and pathology. All procedures with animals were carried out in accordance with Home Office guidelines. Tsc1+/− mice on a BALB/c background (backcross, N ≥ 3; ref. 13) were crossed with Blmm3/m3 mice on a C57BL6 background (12) and Tsc1+/− Blm+/m3 offspring were intercrossed to generate mice for subsequent analyses. DNA was extracted from mouse tail tips using the Qiagen (Chatsworth, CA) QIAamp DNA mini kits and genotyping was done as previously described (12, 13). Necropsy analysis included macroscopic examination of the brain, heart, lungs, kidneys, liver, spleen, and uterus in 20 Tsc1+/− Blm+/m3 and 20 Tsc1+/− Blmm3/m3 mice at 3 to 6 and 9 to 12 months. Half of each organ was fixed and processed into paraffin wax, sectioned at 4 μm, and stained with H&E for microscopic inspection (tissues from 10 mice per genotype were analyzed microscopically at 3-6 and 9-12 months). Estimations of the average number of kidney lesions per mouse were determined as previously described (13). The other half of the organs were snap-frozen in liquid nitrogen–cooled isopentane for laser capture microdissection (LCM).

Immunohistochemistry. Immunohistochemistry using anti-phospho-S6 ribosomal protein (Ser240/244; Cell Signaling Technologies, Beverly, MA) was done as previously described (13).

Somatic mutation analysis. Snap-frozen tissue was sectioned at 10 μm onto PEN (PALM) membrane covered slides and stained with toluidine blue. Tumor and normal tissue was microdissected (PALM LCM) and DNA extracted using the QIAamp DNA micro kits (Qiagen). LOH at the Tsc1 locus was assayed as previously described (13), using wild type/mutant peak ratios of <0.67 or ≥1.3 as indicative of LOH (14). To search for somatic intragenic Tsc1 mutations, the entire Tsc1 open reading frame (ORF) was amplified as 23 fragments (primer sequences are available upon request) and sequenced directly. LOH at the Blm locus was assayed by amplifying and directly sequencing the region encompassing the variant 4109-152 G>A, located in intron 18 of the Blm gene,1

1

Our unpublished data.

using the primers rs8248591/2F 5′-ACAGTTCTGAGAGGGGCTCA-3′ and rs8248591/2R 5′-CCTAGCTTTCCACAGGCACT-3′. LOH of the Blmwt allele was determined by visual inspection of the 4109-152 G>A variant on the sequencing chromatograms by two independent investigators.

Simple sequence length polymorphism markers flanking Tsc1. Six highly polymorphic simple sequence length polymorphism (SSLP) markers flanking Tsc1 (D2Mit484, D2Mit451, D2Mit63, D2Mit37, D2Mit433, and D2Mit431) were analyzed in 18 and 14 tumors with LOH of the Tsc1wt allele from Tsc1+/− Blmm3/m3 and Tsc1+/− Blm+/+ mice, respectively. Amplification of DNA from tumors was done as previously described (13), using a 25-cycle PCR reaction. Two microliters of PCR products were mixed with an ABI GS500 internal size standard and formamide loading buffer and run on an ABI3100 genetic analyzer. Results were analyzed using Genescan v.3.7 software, and LOH was determined as previously described (14).

Statistical analysis. Comparisons of numbers of mice were done using the χ2 test. Lesion counts per mouse were compared using the Mann-Whitney confidence interval tests, and the frequency of LOH was compared using Fisher's exact test.

Blm deficiency has no effect on early Tsc1+/− mortality.Tsc1+/− Blm+/m3 mice were intercrossed. Genotyping showed that of the 360 offspring that survived until weaning, 204 were Tsc1+/− and 134 were wild type (22 failed to be genotyped), indicating that all Tsc1−/− and ∼24% of Tsc1+/− mice died before weaning (P = 0.016). This finding was in agreement with our previous study (13). Genotyping of the Blm allele in Tsc1+/− animals showed no deviation from the expected 1:2:1 ratio (42 Blm+/+, 90 Blm+/m3, 52 Blmm3/m3 mice; 20 failed to be genotyped), indicating that there was no cumulative effect on early mortality for both Tsc1 and Blm deficiency.

Induction of renal tumors in Tsc1+/− mice on a Blm-deficient background. We examined the kidneys of mice at 3 to 6 months by macroscopic inspection and microscopic analysis of five sections ∼200 μm apart. We observed a significant increase in renal tumors in Tsc1+/− mice on a Blm-deficient background; 70% (14 of 20) of Tsc1+/− Blmm3/m3 mice developed macroscopically visible renal lesions (average of 4.2 lesions per mouse) compared with only 15% (3 of 20) of Tsc1+/− Blm+/m3 mice (average of 0.2 lesions per mouse; P = 0.0003; Table 1; Fig. 1). At a microscopic level, Tsc1+/− Blmm3/m3 mice had an average of 23.2 lesions per mouse compared with 5.4 for Tsc1+/− Blm+/m3 mice (P = 0.0203). Renal lesions from Tsc1+/− Blmm3/m3 mice stained positively with anti-phospho-S6 ribosomal protein (Ser240/244; Fig. 1), indicating that these lesions develop through the normal pathway of Tsc-associated tumorigenesis (13).

Table 1.

Frequency of renal lesions in 3- to 6-month-old Tsc1+/− mice on Blm+/m3 and Blmm3/m3 backgrounds

Type of analysisBlm genotype of Tsc1+/− miceTotal no. tumorsMean no. tumors per mouseP
Macroscopic +/m3 0.2 0.0003 
 m3/m3 84 4.2  
Microscopic +/m3 27 5.4 0.0203 
 m3/m3 116 23.2  
Type of analysisBlm genotype of Tsc1+/− miceTotal no. tumorsMean no. tumors per mouseP
Macroscopic +/m3 0.2 0.0003 
 m3/m3 84 4.2  
Microscopic +/m3 27 5.4 0.0203 
 m3/m3 116 23.2  

NOTE: Numbers are based on the analysis of 20 Tsc1+/− mice for macroscopic inspection and 10 Tsc1+/− mice for microscopic inspection.

Figure 1.

Renal cystadenomas in Tsc1+/− mice on a Blm-deficient background. A, an example of a kidney from a Tsc1+/− Blmm3/m3 mouse at 3 to 6 months showing significantly more macroscopically visible lesions compared with (B) a kidney from a Tsc1+/− Blm+/m3 mouse. Microscopic view of cystadenomas from Tsc1+/− Blmm3/m3 mice stained with H&E (C) and immunostained with an anti-phospho-S6 ribosomal protein (p-S6) antibody (D). Region in (D) is a magnified view of the region boxed in (C). Lesions on a Blm-deficient background showed increased expression of phospho-S6 ribosomal protein (brown), suggesting that these lesions develop through the normal pathway of Tsc-associated tumorigenesis. Bar, 5 mm (A-C) and 1 mm (D).

Figure 1.

Renal cystadenomas in Tsc1+/− mice on a Blm-deficient background. A, an example of a kidney from a Tsc1+/− Blmm3/m3 mouse at 3 to 6 months showing significantly more macroscopically visible lesions compared with (B) a kidney from a Tsc1+/− Blm+/m3 mouse. Microscopic view of cystadenomas from Tsc1+/− Blmm3/m3 mice stained with H&E (C) and immunostained with an anti-phospho-S6 ribosomal protein (p-S6) antibody (D). Region in (D) is a magnified view of the region boxed in (C). Lesions on a Blm-deficient background showed increased expression of phospho-S6 ribosomal protein (brown), suggesting that these lesions develop through the normal pathway of Tsc-associated tumorigenesis. Bar, 5 mm (A-C) and 1 mm (D).

Close modal

Somatic mutation analysis of Tsc1 and mechanism of loss of heterozygosity. DNA was extracted from LCM kidney cystadenomas and carcinomas and analyzed for LOH at the Tsc1 locus using an IRES (mutant specific) and exon 8 (wild type specific) quantitative PCR assay, as previously described (13). We observed significantly increased levels of somatic LOH of the Tsc1wt allele in renal lesions from Tsc1+/− Blmm3/m3 mice (53 of 61, 87%) compared with renal lesions from Tsc1+/− Blm+/+ mice (61 of 118, 52%; P < 0.0001; Fig. 2). To determine the mechanism of increased somatic LOH in these lesions, we analyzed the extent of the LOH events using six SSLP markers on mouse chromosome 2, on which Tsc1 resides. Two of 18 tumors showing loss of the Tsc1wt allele from Tsc1+/− Blmm3/m3 mice remained heterozygous at the marker D2Mit431 located ∼5 cM upstream of Tsc1, whereas no tumors with loss of the Tsc1wt allele from Tsc1+/− Blm+/+ mice retained heterozygosity at any of the markers tested. This suggested that the LOH on a Blm-deficient background was mediated by mitotic recombination in agreement with Luo et al. (12). We sought intragenic somatic Tsc1 mutations in DNA from the eight tumors from Tsc1+/− Blmm3/m3 mice that retained the Tsc1wt allele, by direct sequence analysis of the entire Tsc1 ORF. We generated high-quality sequencing for ≥91% of the Tsc1 ORF for each tumor and identified a single intragenic somatic mutation, characterized as a 2-bp deletion in exon 22 (Tsc1 c.2872_2873 delTC). Therefore, in total, we identified ∼89% (54 of 61) of second-hit mutations in Tsc1+/− Blmm3/m3 mice and clearly show that LOH mediated by mitotic recombination is the major mechanism of somatic mutagenesis in these animals. Further studies are warranted to determine the nature of the somatic mutations in the 11% of tumors from Tsc1+/− Blmm3/m3 mice without LOH or coding region variations.

Figure 2.

LOH analysis of renal lesions from Tsc1+/− mice on Blm-proficient (Blm+/+), haploinsufficient (Blm+/m3), and deficient (Blmm3/m3) backgrounds. A, example of LOH in a microdissected renal tumor. Arrow, loss of the wild-type Tsc1 allele in the tumor (bottom) but not in adjacent normal tissue (top). WT, wild-type allele; Mut, mutant allele. Unshaded trace, an ABI GS500 internal size standard. B, % lesions with LOH versus those without LOH from Tsc1+/− Blm+/+, Tsc1+/− Blm+/m3, and Tsc1+/− Blmm3/m3 mice. Ps reflect differences in the frequency of LOH in tumors from Tsc1+/− Blm+/m3 and Tsc1+/− Blmm3/m3 mice compared with Tsc1+/− Blm+/+ mice.

Figure 2.

LOH analysis of renal lesions from Tsc1+/− mice on Blm-proficient (Blm+/+), haploinsufficient (Blm+/m3), and deficient (Blmm3/m3) backgrounds. A, example of LOH in a microdissected renal tumor. Arrow, loss of the wild-type Tsc1 allele in the tumor (bottom) but not in adjacent normal tissue (top). WT, wild-type allele; Mut, mutant allele. Unshaded trace, an ABI GS500 internal size standard. B, % lesions with LOH versus those without LOH from Tsc1+/− Blm+/+, Tsc1+/− Blm+/m3, and Tsc1+/− Blmm3/m3 mice. Ps reflect differences in the frequency of LOH in tumors from Tsc1+/− Blm+/m3 and Tsc1+/− Blmm3/m3 mice compared with Tsc1+/− Blm+/+ mice.

Close modal

Do Tsc1+/− Blm+/m3 mice show increased tumor burden? Goss et al. (15) showed that mice heterozygous for a targeted null mutation of Blm developed lymphoma earlier than wild-type littermates when challenged with murine leukemia virus and developed twice the number of intestinal tumors when crossed with Apc+/Min mice. Furthermore, Gruber et al. (16) showed that human carriers of a BLM mutation have an increased risk for colorectal cancer. In contrast, we did not observe significantly more renal lesions in Tsc1+/− Blm+/m3 mice at 3 to 6 months when compared with Tsc1+/− Blm+/+ mice that had previously been backcrossed (N > 3) onto two different backgrounds (BALB/c and C57BL6; ref. 13). However, we did observe significantly increased levels of somatic LOH of the Tsc1wt allele in renal lesions from Tsc1+/− Blm+/m3 mice (23 of 32, 71%; P = 0.03). To determine whether tumors from Tsc1+/− Blm+/m3 mice had lost the Blmwt allele, we analyzed the polymorphism 4109-152 G>A, which lies within intron 18 of the Blm gene. Nineteen of 20 (95%) renal lesions from Tsc1+/− Blm+/m3 mice that had lost the Tsc1wt allele and were informative for this marker retained the Blmwt allele, indicating that the increased somatic LOH at the Tsc1 locus in these mice was mediated by Blm haploinsufficiency.

McDaniel et al. (17) recently showed that the Blmm3 allele is in fact hypomorphic and expresses a low level of the Blm protein. This may explain our failure to find significantly increased tumor burden in Tsc1+/− Blm+/m3 mice. However, our results indicate that the previously reported tumor predisposition associated with Blm haploinsufficiency (15, 16) may be mediated by increased somatic LOH.

Does the nature of the second hit influence the spectrum and distribution of Tsc-associated lesions? We examined Tsc1+/− Blm+/m3 and Tsc1+/− Blmm3/m3 mice at 9 to 12 months and compared these with Tsc1+/− Blm+/+ mice to see whether the increased frequency of LOH at Tsc1 in the Blm-deficient animals influenced the spectrum and distribution of their tumors. Although Blmm3/m3 mice are prone to a wide variety of cancers, the incidence of lesions is very low (3%) before 1 year (12). Renal tumors were again more frequent in Tsc1+/− Blmm3/m3 mice compared with Tsc1+/− Blm+/m3 mice (P = 0.0027). Consistent with the increased numbers of tumors noted at 3 to 6 months, Tsc1+/− Blmm3/m3 mice had a higher proportion of more advanced renal tumors (with a solid to cystic tumor ratio of 1:5) compared with Tsc1+/− Blm+/m3 mice (ratio of 1:11.5), with a single renal lesion metastasizing to the lungs.

The spectrum and distribution of extrarenal tumors in Tsc1+/− mice on Blm+/m3 or Blmm3/m3 backgrounds was not significantly different to Tsc1+/− mice on a Blm+/+ background. Only an isolated liver hemangioma was observed in a single Tsc1+/− Blmm3/m3 mouse, and no other lesions were seen after macroscopic and microscopic examination of the brain, heart, lungs, spleen, and uterus. Renal lesions have also been induced in other Tsc1+/− mice (18) by using chemical mutagens, which cause a high frequency of somatic point mutations on the wild-type allele, and consistent with our observations, these animals do not seem to develop any unusual extrarenal tumors. These data indicate that the nature of the second hit does not affect the spectrum and distribution of extrarenal lesions in Tsc1+/− mice.

In conclusion, our study confirms the use of the Blmm3/m3 mouse for inducing tumorigenesis in cancer-prone mouse models. The high levels of LOH in the resultant tumors (and consequently low level of intragenic mutations) will help facilitate the mapping of loci that may play a role in tumor progression.

Grant support: Tenovus, Tuberous Sclerosis Association, Tuberous Sclerosis Alliance, and Wales Gene Park.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank A. Bradley for the Blmm3/m3 mice and probe for genotyping; D.F.R. Griffiths, A.J.H. Smith, A. Clarke, and K. Davies for helpful advice; and K. Allsopp, J. Dunstan, R. Davies, S. Macdonald, and D. Adams for technical support.

1
Hickson ID. RecQ helicases: caretakers of the genome.
Nat Rev Cancer
2003
;
3
:
169
–78.
2
Ellis NA, Groden J, Ye TZ, et al. The Bloom's syndrome gene product is homologous to RecQ helicases.
Cell
1995
;
83
:
655
–66.
3
Karow JK, Constantinou A, Li J-L, West SC, Hickson ID. The Bloom's syndrome gene product promotes branch migration of Holliday junctions.
Proc Natl Acad Sci U S A
2000
;
97
:
6504
–8.
4
Sun H, Karow JK, Hickson ID, Maizels N. The Bloom's syndrome helicase unwinds G4 DNA.
J Biol Chem
1998
;
273
:
27587
–92.
5
Langland G, Elliott J, Li Y, Creaney J, Dixon K, Groden J. The BLM helicase is necessary for normal DNA double-strand break repair.
Cancer Res
2002
;
62
:
2766
–70.
6
German J. Bloom syndrome: a Mendelian prototype of somatic mutational disease.
Medicine (Baltimore)
1993
;
72
:
393
–406.
7
Groden J, German J. Bloom's syndrome. XVIII. Hypermutability at a tandem-repeat locus.
Hum Genet
1992
;
90
:
360
–7.
8
Ellis NA, Proytcheva M, Sanz MM, Ye TZ, German J. Transfection of BLM into cultured bloom syndrome cells reduces the sister-chromatid exchange rate toward normal.
Am J Hum Genet
1999
;
65
:
1368
–74.
9
Bahr A, De Graeve F, Kedinger C, Chatton B. Point mutations causing Bloom's syndrome abolish ATPase and DNA helicase activities of the BLM protein.
Oncogene
1998
;
17
:
2565
–71.
10
Yusa K, Horie K, Kondoh G, et al. Genome-wide phenotype analysis in ES cells by regulated disruption of Bloom's syndrome gene.
Nature
2004
;
429
:
896
–9.
11
Guo G, Wang W, Bradley A. Mismatch repair genes identified using genetic screens in Blm-deficient embryonic stem cells.
Nature
2004
;
429
:
891
–5.
12
Luo G, Santoro IM, McDaniel LD, et al. Cancer predisposition caused by elevated mitotic recombination in Bloom mice.
Nat Genet
2000
;
26
:
424
–9.
13
Wilson C, Idziaszczyk S, Parry L, et al. A mouse model of tuberous sclerosis 1 showing background specific early post-natal mortality and metastatic renal cell carcinoma.
Hum Mol Genet
2005
;
14
:
1839
–50.
14
Zhou CZ, Qiu GQ, Zhang F, He L, Peng ZH. Loss of heterozygosity on chromosome 1 in sporadic colorectal carcinoma.
World J Gastroenterol
2004
;
10
:
1431
–5.
15
Goss KH, Risinger MA, Kordich JJ, et al. Enhanced tumor formation in mice heterozygous for Blm mutation.
Science
2002
;
297
:
2051
–3.
16
Gruber SB, Ellis NA, Scott KK, et al. BLM heterozygosity and the risk of colorectal cancer.
Science
2002
;
297
:
2013
.
17
McDaniel LD, Chester N, Watson M, Borowsky AD, Leder P, Schultz RA. Chromosome instability and tumor predisposition inversely correlate with BLM protein levels.
DNA Repair (Amst)
2003
;
2
:
1387
–404.
18
Kobayashi T, Minowa O, Sugitani Y, et al. A germ-line Tsc1 mutation causes tumor development and embryonic lethality that are similar, but not identical to, those caused by Tsc2 mutation in mice.
Proc Natl Acad Sci U S A
2001
;
98
:
8762
–7.